npj Computational Materials最新文献

{"title":"Topology-aware E(3)-equivariant learning for physically consistent piezoelectric tensor prediction","authors":"Ruihan Liu,Jianbo Yu,Yu Ji,Hang Ruan,Xiaofeng Yang","doi":"10.1038/s41524-026-02331-3","DOIUrl":"https://doi.org/10.1038/s41524-026-02331-3","url":null,"abstract":"","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"199 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149049610","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Polarization-controlled ultrafast carrier dynamics in sliding ferroelectrics revealed by machine learning-accelerated nonadiabatic dynamics simulations","authors":"Dong-Hui Xu,Shuangli Yue,Yitong Wang,Ronghui Tang,Xiang-Yang Liu,Li Zhang,Yunpeng Lu,Xian Wang,Mingli Yang","doi":"10.1038/s41524-026-02348-8","DOIUrl":"https://doi.org/10.1038/s41524-026-02348-8","url":null,"abstract":"","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"14 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149049607","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unveiling inelastic phonon scattering at Al-Al2O3 interfaces by machine-learning molecular dynamics","authors":"Jinyuan Xu,Yangyu Guo","doi":"10.1038/s41524-026-02326-0","DOIUrl":"https://doi.org/10.1038/s41524-026-02326-0","url":null,"abstract":"","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"95 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149021779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"OMC-bench and AtomBit-OMC: task-aligned benchmarks and robust, interpretable machine-learned interatomic potentials for organic molecular crystals","authors":"Muyu Lu,Chengxi Zhao,Siyu Hu,Danyang Chen,Chen Wang,Yixi Zhang,Yaolong Zhang,Man Luo,Fan Yu,Zhuoying Zhu,Bin Jiang,Guangming Tan,Weile Jia,Jun Jiang,Linjiang Chen","doi":"10.1038/s41524-026-02315-3","DOIUrl":"https://doi.org/10.1038/s41524-026-02315-3","url":null,"abstract":"","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"9 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149021782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Practical and accurate density functionals for computational heterogeneous catalysis","authors":"Benjamin X. Shi, Timothy C. Berkelbach","doi":"10.1038/s41524-026-02320-6","DOIUrl":"https://doi.org/10.1038/s41524-026-02320-6","url":null,"abstract":"Density functional theory underpins modern atomistic simulations of transition-metal surfaces. It can predict key properties linked to catalytic performance, such as adsorption energies and barrier heights, enabling new paradigms in rational catalyst design. These applications require reliable density functionals, achieving errors below 13 kJ mol−1, sometimes called transition-metal chemical accuracy, which remains challenging. We introduce a non-self-consistent approach to lower the cost of higher-rung functionals by orders of magnitude, enabling application to larger and more complex metal surfaces than their self-consistent counterparts. Building on this, we develop a new hybrid and double-hybrid functional that achieves better accuracy than standard semilocal functionals, with the double-hybrid functional reaching transition-metal chemical accuracy on average across 39 experimental adsorption reactions. In addition, they both demonstrate balanced performance for 17 barrier heights and correct qualitative failures of standard functionals, including CO adsorption on Pt(111) and graphene on Ni(111). These functionals can be readily implemented in existing density functional theory codes, and we provide code for their integration into open-source high-throughput workflows to facilitate broad adoption. More broadly, this framework provides a systematic route towards improved functionals for heterogeneous catalysis and complex materials.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"1 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149010597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exceptionally high carrier mobility in hexagonal diamond","authors":"Zirui He, Shang-Peng Gao, Meng Chen","doi":"10.1038/s41524-026-02330-4","DOIUrl":"https://doi.org/10.1038/s41524-026-02330-4","url":null,"abstract":"Hexagonal diamond (h-diamond), or Lonsdaleite, is a promising wide-bandgap semiconductor known for its high thermal conductivity and hardness. Based on ab initio calculations, we demonstrate its exceptionally high carrier mobilities. At room temperature, the hole mobilities along the ⊥c and ∥c directions are 6000 and 6024 cm2V−1s−1, respectively, while the corresponding electron mobilities reach 12,339 and 28,473 cm2V−1s−1. These values are significantly superior to those of most known semiconductors, including cubic diamond. The small effective masses in h-diamond are comparable to those in the cubic phase, which cannot explain its substantially higher mobilities. Instead, two underlying mechanisms are uncovered. First, selection rules enforced by the symmetry of h-diamond significantly suppress scattering, particularly for transverse acoustic phonons, which predominate in the cubic phase around room temperature. Secondly, the spatial mismatch between the electronic wavefunctions and phonon-induced scattering potentials leads to real-space electron-phonon decoupling, which manifests as the suppression of out-of-plane polarised longitudinal acoustic scattering for holes, and a systematic weakening of acoustic scattering for electrons.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"47 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149010596","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unsupervised deep learning for limited-angle STEM-EDX tomography: application to 3D chemical analysis of phase-change memory devices","authors":"Daniel del Pozo Bueno, Serge Brosset, Theo Monniez, Gabriele Navarro, Philippe Ciuciu, Zineb Saghi","doi":"10.1038/s41524-026-02325-1","DOIUrl":"https://doi.org/10.1038/s41524-026-02325-1","url":null,"abstract":"Energy-dispersive X-ray (EDX) tomography in scanning transmission electron microscopy (STEM) enables three-dimensional elemental mapping at the nanoscale. However, its application remains challenging when sample geometry severely restricts the accessible tilt range, and the electron dose must be limited to prevent beam damage. Limited-angle acquisition produces missing-wedge artifacts, including elongation and anisotropic resolution, while low-dose measurements further degrade reconstruction quality and hinder reliable quantification. Here, we introduce an unsupervised deep-learning framework combining a Deep Image Prior with total-variation regularization (DIP-TV) for STEM-EDX tomography under extremely limited-angle conditions. We further propose a multi-channel formulation, DIPm-TV, that jointly reconstructs multiple elemental volumes by exploiting their spatial correlations. Using a synthetic three-channel phantom, we show that DIPm-TV substantially reduces artifacts arising from severe angular limitations in the presence of moderate noise, outperforming the simultaneous iterative reconstruction technique and compressed-sensing-based approaches. We then apply DIPm-TV to two Ge–Sb–Te-based memory devices: an as-fabricated virgin device and a device programmed into the SET state, both prepared as conventional cross-sectional focused-ion-beam lamellae. Under a severely limited tilt range (±40°) and a low-dose ((2.0times {10}^{5}{e}^{-}{text{AA }}^{-2})), DIPm-TV yields reliable voxel-by-voxel 3D elemental maps from EDX signals alone, without external structural priors such as HAADF-STEM images. The reconstructed volumes exhibit near-isotropic spatial resolution and reveal compositional heterogeneities associated with device operation. This approach enables 3D chemical characterization of semiconductor devices in experimentally accessible sample geometries for which conventional reconstruction methods are strongly compromised by angular limitations.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"31 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149010599","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantum-inspired inverse design of heterogeneous catalysts for hydrogen evolution reaction","authors":"Sangro Lee, Chang Woo Kim, Hyun Woo Kim","doi":"10.1038/s41524-026-02324-2","DOIUrl":"https://doi.org/10.1038/s41524-026-02324-2","url":null,"abstract":"Data-driven discovery of stable and functional materials remains a major challenge in chemistry and materials science, both from classical and quantum perspectives. Quantum machine learning provides new opportunities to optimize materials across vast chemical spaces, but hardware limitations restrict its practical application. Classical models can efficiently process large datasets, but their ability to explore unseen chemical spaces is fundamentally limited by the uncertainty of out-of-distribution predictions. Although transfer learning enables exploration beyond the training distribution, fine-tuning on the target domain often distorts the latent representations learned from the source domain. To address these challenges, we introduce a quantum-inspired generative framework for discovering catalytic materials for hydrogen evolution reactions (HER). Inspired by variational quantum circuits (VQCs), which outperform classical models in certain tasks, we hypothesized that the norm-conserving property of VQCs plays a key role in materials representation and optimization. Thus, we incorporated orthogonal transformations into the classical generative model to maintain latent space structure and enhance materials discovery. With additional data through transfer learning, the classical model with orthogonal transformations becomes useful. Our results provide practical guidelines for applying quantum machine learning to materials discovery and suggest the potential of quantum-inspired generative models to explore previously inaccessible chemical spaces.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"475 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149010598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Physics-guided diffusion models for inverse design of disordered metamaterials","authors":"Ziyuan Xie, Weipeng Xu, Dazhi Zhao, Wenchang Zhang, Daoyang Dong, Bingbing Xu, Ning Liu, Sheng Mao, Tianju Xue","doi":"10.1038/s41524-026-02328-y","DOIUrl":"https://doi.org/10.1038/s41524-026-02328-y","url":null,"abstract":"Disordered metamaterials are promising for programming physical properties, yet their inverse design remains challenging due to the non-intuitive structure-property relationships. Recent generative approaches, particularly diffusion models, have shown potential in high-dimensional inverse design tasks. However, existing methods typically rely on task-specific training strategies, such as conditional data-driven or physics-informed loss functions. Consequently, models must be retrained from scratch whenever governing equations, boundary conditions, or design objectives change, limiting their flexibility and generalization. In this work, we propose physics-guided diffusion models that leverage differentiable physics-based solvers to instantly guide the generative process for inverse design. Drawing inspiration from classifier guidance, we develop a sampling strategy that directly incorporates physics guidance into the reverse stochastic differential equations. Using gradients from differentiable solvers, our approach enables task-adaptive generation within a learned morphology prior, while requiring the diffusion model to be trained only once on unlabeled data. Focusing on 2D disordered closed-cell foam metamaterials, we present three design tasks: (1) achieving target effective thermal conductivity, (2) matching desired load-displacement response, and (3) maximizing energy absorption involving fractures. The results in each scenario demonstrate the versatility, efficiency, and practicality of physics-guided diffusion models for tackling complex inverse design problems in disordered metamaterials and beyond.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"31 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149010504","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Beyond the “source-sink” paradigm: catalytic mechanism of low-angle tilt grain boundaries in vacancy generation and annihilation","authors":"Hao Wang, YongQuan Wu, JingLin You","doi":"10.1038/s41524-026-02329-x","DOIUrl":"https://doi.org/10.1038/s41524-026-02329-x","url":null,"abstract":"Grain boundary (GB)-vacancy interactions play a pivotal role in tailoring material properties. However, the prevailing paradigm that “GBs function as vacancy sources/sinks” lacks robust atomic-scale evidence, thereby failing to discriminate the intrinsic mechanisms governing vacancy dynamics and leaving critical unresolved issues—most notably the GB capacity for vacancy generation/annihilation. To fill this gap, we adopt molecular dynamics simulations to systematically observe the vacancy thermal equilibration (VTE) process, wherein low-angle tilt GBs (LATGBs) are engineered as the exclusive vacancy sources and sinks. Our results unveil a remarkable phenomenon: LATGBs maintain invariant atomic structures despite continuous massive vacancy absorption and emission throughout VTE. Accordingly, we propose that LATGBs function as catalysts for vacancy generation/annihilation, with unlimited vacancy absorption/emission capacity. Furthermore, we corroborate four critical atomic-scale findings: (1) Vacancy generation/annihilation rate at edge dislocations reaches 8.62 ns⁻¹·nm⁻¹; (2) Atomic migration rate within dislocation vacant channels is at least three orders of magnitude faster than grain interior diffusion; (3) Seven of eight sites surrounding each vacant channel are catalytically active due to ultralow vacancy formation energies; (4) Compressive-tensile stress interfaces along dislocations mediate vacancy migration between active sites and grain interiors. These atomic-scale insights establish a complete catalytic cycle governing LATGB-mediated vacancy generation/annihilation.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"433 1 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2026-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149010506","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}